Remote Asset Analytics Engine for Cross-Site Model Variations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Firms using remote monitoring systems face inefficiencies and high costs due to the need for customized analytics caused by imperfect asset models, which vary in sensor parameters and specifications across different sites, making it difficult to apply standard analytics and compare results across a fleet of assets.
Innovation Solution
An analytic system that includes a processor to retrieve expressions and arguments, attach time-series parameters, convert data to standard units, perform calculations, and write results, allowing for standardized analytics across assets with varying configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If customized analytics are created for each asset, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent implements a standardized analytics engine that can process multiple asset types with different sensor configurations using the same calculation framework. The system uses parameter tags and expression templates that can be universally applied across diverse assets, eliminating the need to create separate customized analytics for each asset while maintaining measurement precision through configurable parameters.
Solution Approach 2:
The system allows dynamic configuration of asset parameters such as sensor types, units of measurement, and calculation expressions. By changing parameters rather than creating entirely new analytics models, the system adapts to different asset configurations while using the same underlying analytics engine, thus reducing complexity while maintaining precision.
2Device complexity
If standardized analytics are applied across all assets, then device complexity is reduced, but measurement precision deteriorates due to imperfect asset models
Solution Approach 1:
The patent applies local quality by allowing asset-specific parameter configurations within the standardized analytics framework. Each asset can have its own sensor parameters, units, and calculation expressions tailored to its specific characteristics, while still using the same overall analytics engine. This enables standardized processing with localized precision adjustments.
Solution Approach 2:
The system introduces an intermediary layer of parameter tags and expression templates that mediate between the standardized analytics engine and the diverse asset models. This intermediary layer translates various asset-specific parameters into a common calculation framework, preserving measurement precision while enabling standardization.
3Measurement precision
If asset models are made more detailed to capture variations, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments asset models into modular parameter tags that can be independently configured and combined. Instead of creating monolithic complex asset models, the system breaks down asset characteristics into discrete, configurable parameters that can be selectively applied, reducing overall model complexity while maintaining the ability to represent detailed variations when needed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An analytic system comprising a plurality of asset models for modeling a plurality of assets to be monitored at remote sites (4a, 4b) using a plurality of sensors using different original units of measurement, the plurality of asset models comprising an expression for a calculation for the plurality of assets; a memory comprising computer instructions; and a processor coupled to the memory, wherein, when executing the computer instructions, the processor effectuates operations to perform analytics on the plurality of asset models. The operations comprise for each of the plurality of assets: retrieving the expression and one or more arguments for a calculation of an asset model of the plurality of asset models, the asset model representing an asset at a remote site; attaching time-series parameters associated with the asset and sensed by a sensor at a remote site, the time-series parameters being attached as required to the arguments, wherein the time-series parameters are specified by one or more tags, wherein a tag is the primary key identifier used to uniquely identify each time-series parameter; wherein the analytic system comprises a tag pattern that matches tags appearing in the modeling of different assets of the plurality of assets; retrieving time-series data based on the attaching step; converting the time-series data from an original unit associated with the time-series data to a standard unit; performing a calculation based on the expression and the converted time-series data to achieve a result, wherein performing the calculation comprises using the tag pattern in a calculation in which there are differences in configuration of the plurality of assets; converting the result to the original unit; and writing the result to a tag associated with the one or more arguments, wherein writing the result to the tag is performed by writing the result to the tag using the original unit; the analytic system being configured to remotely monitor the plurality of assets based on the performed analytics.